Abstract: Most spatial clustering methods utilize fixed thresholds in the process of clustering which assume homogeneous (or even) distribution of the spatial points rather than inhomogeneous (or uneven) scattering. However, spatial points usually distribute unevenly (in different density) which makes the fixed threshold methods inappropriate. Thus, an adaptive density-based spatial cluster algorithm, ADBSC for short, is developed in this paper. A new measurement of spatial local density, named as maximum distance in k-spatial nearest neighborhood (k-NN for short), is proposed. In k-NN a Dap (distance alternation proportion) threshold is used to judge whether a point is a core, and check whether density is directly reachable or reachable. All density-reachable cores and their boundary points compose a spatial cluster so that the spatial cluster is implemented, which adapts to changes of local density among spatial points. Furthermore, the ADBSC algorithm is described in detail. A simulation test demonstrates that the ADBSC algorithm is capable to discovery arbitrary-shape clusters and is robust for noises. A comparison between ADBSC and DBSCAN shows that the ADBSC is more flexible and efficient than the DBSCAN. Finally, the real-life data is employed to prove that the ADBSC has more practicality than DBSCAN.